PROJECT SUMMARY Diabetic kidney disease (DKD) is the most common cause of end-stage renal disease in the US, with limited treatment options. Several recent clinical trials have not succeeded, likely owing to the lack of early intervention. Glomerular endothelial cell (GEC) dysfunction promotes the progression of DKD, and evidence suggests that GEC injury is an early event in diabetic kidneys that may precede the development of microalbuminuria. However, the mechanisms of early GEC injury in DKD remain unclear. Thus, better understanding of the underlying processes of GEC injury is urgently required for development of early therapeutic intervention. To date, the study of GEC injury in DKD has relied on the analysis of whole kidney tissues or isolated glomeruli, which provides indirect and limited information. Recently, we developed and optimized a novel method of effective isolation of GECs from transgenic mice expressing enhanced yellow fluorescent protein (EYFP) under the endothelium-specific Flk1 promoter. Using this approach, we compared the transcriptomic profiles of GECs isolated from streptozotocin (STZ)-induced diabetic endothelial nitric oxide synthase (eNOS)-null mice with GECs from nondiabetic controls. Many of the differentially expressed genes in diabetic GECs were involved in the regulation of angiogenesis and endothelial injury. Among these, a secreted glycoprotein, leucine-rich alpha-2 glycoprotein (LRG1) was found to be highly upregulated in diabetic GECs. LRG1 was recently shown to be the necessary co-activator of TGF-? signaling that is essential in neoangiogenesis of retinal and cancer cells. Thus, we explored whether LRG1 has a regulatory function in angiogenesis in early DKD.We found that 1) mRNA expression of LRG1 is significantly increased in GECs of diabetic mice, as well as in glomeruli of DKD patients; 2) correspondingly, LRG1 expression was found predominantly in GECs of mouse and human kidneys; 3) LRG1 protein is increased in serum and urine samples of DKD patients; 4) high glucose induces LRG1 expression in glomeruli and GECs cultured in vitro; 5) LRG1 knockout reduces albuminuria and glomerular injury, and attenuates angiogenesis in diabetic mice; and 6) LRG1 promotes angiogenesis via induction of the non-canonical TGF-? pathway in GECs, and knockdown of LRG1 in GECs is sufficient to diminish angiogenesis in vitro. Based on these salient findings, we hypothesize that LRG1 is a key pathogenic contributor of disease progression in early DKD by increasing angiogenesis and injury in GECs. Therefore, we propose to 1) determine the role of LRG1 in DKD using type 1 and type 2 DM mouse models; 2) determine the EC-specific role of LRG1 in early and late stages of DKD; and 3) determine the molecular mechanism of LRG1-induced diabetic GEC injury in vitro. Results from the proposed study will build the strong foundation for the future clinical studies to determine whether LRG1 is a potential biomarker of early DKD and whether the systemic blockade of LRG1 would offer therapeutic benefit to intervene against DKD progression.